Hi there,

 

I think this may be inappropriate for this list as I searched the archive and all the messages of this type were pointed to e-mail to '[log in to unmask]'.  I have e-mailed it to that list.  But, I’ve sent it here because it may be such a common problem, that one of you could show me the solution and it wouldn’t take but a few minutes of your time.  I would sincerely appreciate a quick glance.

 

I’m trying to fit a Hierarchical Model that performs a meta-analysis across study type but accounting for study type.  There are seven studies, 2 that are randomized and 5 that are not.  The data in each of the studies is the observed deaths (rmortRT) out of the total number patients (nmortRT).  The study design information is held in the variable DesignRT.  DesignRT=1 is a randomized design and DesignRT=2 is a non-randomized design.

 

Each study design type is expected to have a mean death rate after logit transformation, muRT[DesignRT[i]].  The overall population mean death rate (combining over study type) after logit transformation is munotRT.  

 

When I compile the model and the data, I get a multiple definitions of node muRT[2] message.  Why isn’t this working?  This project is due next week so any help that you could give me in the next day or so would be greatly appreciated.  I’m sort of at a standstill. 

 

Take care,

Teresa

 

 

model{ for (i in 1:Nstudy){                          

                                     rmortRT[i] ~ dbin(pRT[i], nmortRT[i])

                                     logit(pRT[i]) <-  RT[i]

                                     RT[i] ~ dnorm(muRT[DesignRT[i]], tauRT[DesignRT[i]])                             

                                     useless[i] <- StudyRT[i] + StentNumRT[i]+ gpiiaiiibRT[i]

                                     muRT[DesignRT[i]] ~ dnorm(munotRT, taunotRT)

}

                                                                       

tauRT[1] <- pow(sigmaRT[1], -2)

tauRT[2] <- pow(sigmaRT[2], -2)

sigmaRT[1] ~ dunif(0,100)

sigmaRT[2] ~ dunif(0, 100)

ppostRT[1] <- exp(muRT[DesignRT[1]])/(1 + exp(muRT[DesignRT[1]]))

ppostRT[2] <- exp(muRT[DesignRT[2]])/(1 + exp(muRT[DesignRT[2]]))

 

munotRT ~ dnorm(0, 0001)

 

taunotRT <-pow(sigmanotRT, -2)

 

sigmanotRT ~ dunif(0,100)

 

ppostRTnot <- exp(munotRT)/(1 + exp(munotRT))

 

}

 

#DATA AJ STUDIES

list(Nstudy=7)

                                                                                                            

StudyRT[]               DesignRT[]             StentNumRT[]         gpiiaiiibRT[]           nmortRT[]               rmortRT[]

1              1              1              1              50            0

2              2              0.913       0.87         46            2

3              2              NA           NA           80            3

4              2              0.39         0              31            0

5              2              0.67         0.262       70            5

6              1              0.93         0.95         240          11

7              2              1              0.88         52            1

END

 

 

Teresa Nelson, M.S.

Principal Statistician

Princeton Reimbursement Group

9801 Dupont Avenue South - Suite 295

Minneapolis, MN 55431

Phone:  (952) 345-6413

Fax:      (715) 755-2767

e-mail:   [log in to unmask]